Object Detection Demo

Welcome to the object detection inference walkthrough! This notebook will walk you step by step through the process of using a pre-trained model to detect objects in an image. Make sure to follow the installation instructions before you start.

Imports

In [1]:
import glob
import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import time
import zipfile

from collections import defaultdict
from io import StringIO
from matplotlib import pyplot as plt
from PIL import Image

# if tf.__version__ != '1.4.0':
#   raise ImportError('Please upgrade your tensorflow installation to v1.4.0!')
/home/ubuntu/anaconda3/lib/python3.6/importlib/_bootstrap.py:205: RuntimeWarning: compiletime version 3.5 of module 'tensorflow.python.framework.fast_tensor_util' does not match runtime version 3.6
  return f(*args, **kwds)
In [ ]:
tf.__version__

Env setup

In [2]:
# This is needed to display the images.
%matplotlib inline

# This is needed since the notebook is stored in the object_detection folder.
sys.path.append("..")

Object detection imports

Here are the imports from the object detection module.

In [3]:
from utils import label_map_util

from utils import visualization_utils as vis_util

Model preparation

Variables

Any model exported using the export_inference_graph.py tool can be loaded here simply by changing PATH_TO_CKPT to point to a new .pb file.

By default we use an "SSD with Mobilenet" model here. See the detection model zoo for a list of other models that can be run out-of-the-box with varying speeds and accuracies.

In [4]:
# What model to download.
# MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_17'
# MODEL_FILE = MODEL_NAME + '.tar.gz'
# DOWNLOAD_BASE = 'http://download.tensorflow.org/models/object_detection/'

# Path to frozen detection graph. This is the actual model that is used for the object detection.
# PATH_TO_CKPT = MODEL_NAME + '/frozen_inference_graph.pb'

# PATH_TO_CKPT = '/home/ubuntu/homeroot/carnd-capstone-repo/classifier/faster_rcnn_inception_v2_real.pb'
PATH_TO_CKPT = '/home/ubuntu/homeroot/carnd-capstone-repo/classifier/faster_rcnn_inception_v2_sim.pb'
# PATH_TO_CKPT = '/home/ubuntu/homeroot/carnd-capstone-repo/classifier/ssd_inception_v2_real.pb'
# PATH_TO_CKPT = '/home/ubuntu/homeroot/carnd-capstone-repo/classifier/ssd_inception_v2_sim.pb

# List of the strings that is used to add correct label for each box.
# PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')

PATH_TO_LABELS = '/home/ubuntu/homeroot/carnd-capstone-train/data/udacity_label_map.pbtxt'

NUM_CLASSES = 4

Download Model

In [ ]:
# opener = urllib.request.URLopener()
# opener.retrieve(DOWNLOAD_BASE + MODEL_FILE, MODEL_FILE)
# tar_file = tarfile.open(MODEL_FILE)
# for file in tar_file.getmembers():
#   file_name = os.path.basename(file.name)
#   if 'frozen_inference_graph.pb' in file_name:
#     tar_file.extract(file, os.getcwd())

Load a (frozen) Tensorflow model into memory.

In [5]:
detection_graph = tf.Graph()
with detection_graph.as_default():
  od_graph_def = tf.GraphDef()
  with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
    serialized_graph = fid.read()
    od_graph_def.ParseFromString(serialized_graph)
    tf.import_graph_def(od_graph_def, name='')

Loading label map

Label maps map indices to category names, so that when our convolution network predicts 5, we know that this corresponds to airplane. Here we use internal utility functions, but anything that returns a dictionary mapping integers to appropriate string labels would be fine

In [6]:
label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_map_util.create_category_index(categories)

Helper code

In [7]:
def load_image_into_numpy_array(image):
  (im_width, im_height) = image.size
  return np.array(image.getdata()).reshape(
      (im_height, im_width, 3)).astype(np.uint8)

Detection

In [8]:
# For the sake of simplicity we will use only 2 images:
# image1.jpg
# image2.jpg
# If you want to test the code with your images, just add path to the images to the TEST_IMAGE_PATHS.
# PATH_TO_TEST_IMAGES_DIR = 'test_images'
# TEST_IMAGE_PATHS = [ os.path.join(PATH_TO_TEST_IMAGES_DIR, 'image{}.jpg'.format(i)) for i in range(1, 3) ]

# TEST_IMAGE_PATHS = \
#     glob.glob('/home/ubuntu/homeroot/_downloads/anthony-sarkins/real_training_data/*/*.jpg')[::30] + \
#     glob.glob('/home/ubuntu/homeroot/_downloads/alex-lechner/udacity_testarea_rgb/*jpg')[::30]

TEST_IMAGE_PATHS = \
    glob.glob('/home/ubuntu/homeroot/_downloads/anthony-sarkins/sim_training_data/sim_data_capture/*.jpg')[::30] + \
    glob.glob('/home/ubuntu/homeroot/_downloads/alex-lechner/simulator_dataset_rgb/*/*.jpg')[::30]

def batches(l, n):
    for i in range(0, len(l), n):
        yield l[i:i + n]

# Size, in inches, of the output images.
IMAGE_SIZE = (15, 10)
In [9]:
with detection_graph.as_default():
    with tf.Session(graph=detection_graph) as sess:
        # Definite input and output Tensors for detection_graph
        image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
        
        # Each box represents a part of the image where a particular object was detected.
        detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
        
        # Each score represent how level of confidence for each of the objects.
        # Score is shown on the result image, together with the class label.
        detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
        detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
#         num_detections = detection_graph.get_tensor_by_name('num_detections:0')
        
        batch_size = 1
    
        for test_image_paths_batch in batches(TEST_IMAGE_PATHS, batch_size):
            images_np = []
        
            for image_path in test_image_paths_batch:
                image = Image.open(image_path)
        
                # the array based representation of the image will be used later in order to prepare the
                # result image with boxes and labels on it.
                image_np = load_image_into_numpy_array(image)
    
                images_np.append(image_np)
    
            images_np = np.array(images_np)
    
            # Actual detection.
            start_time = time.time()
            (boxes_all, scores_all, classes_all) = sess.run(
                [detection_boxes, detection_scores, detection_classes],
                feed_dict={image_tensor: images_np})
            duration = time.time() - start_time
            print('duration over {} images: {}'.format(batch_size, duration))

            for image_np, boxes, scores, classes in zip(images_np, boxes_all, scores_all, classes_all):
                # Visualization of the results of a detection.
                vis_util.visualize_boxes_and_labels_on_image_array(
                    image_np,
                    np.squeeze(boxes),
                    np.squeeze(classes).astype(np.int32),
                    np.squeeze(scores),
                    category_index,
                    use_normalized_coordinates=True,
                    line_thickness=8)
                plt.figure(figsize=IMAGE_SIZE)
                plt.imshow(image_np)
duration over 1 images: 3.030214548110962
duration over 1 images: 0.11785578727722168
duration over 1 images: 0.11716461181640625
duration over 1 images: 0.11758685111999512
duration over 1 images: 0.11779022216796875
duration over 1 images: 0.11783146858215332
duration over 1 images: 0.11999297142028809
duration over 1 images: 0.11763453483581543
duration over 1 images: 0.11805105209350586
duration over 1 images: 0.1173391342163086
duration over 1 images: 0.11778640747070312
duration over 1 images: 0.11748743057250977
duration over 1 images: 0.11733460426330566
duration over 1 images: 0.11745023727416992
duration over 1 images: 0.11754536628723145
duration over 1 images: 0.11757397651672363
duration over 1 images: 0.1173257827758789
duration over 1 images: 0.11746406555175781
duration over 1 images: 0.11719346046447754
duration over 1 images: 0.11703634262084961
duration over 1 images: 0.11769223213195801
/home/ubuntu/anaconda3/lib/python3.6/site-packages/matplotlib/pyplot.py:524: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).
  max_open_warning, RuntimeWarning)
duration over 1 images: 0.1173391342163086
duration over 1 images: 0.11734223365783691
duration over 1 images: 0.11713123321533203
duration over 1 images: 0.117401123046875
duration over 1 images: 0.11653447151184082
duration over 1 images: 0.11904335021972656
In [ ]: